Determining Ownership in Multiple-Grantor Trusts
In this article for the New York Law Journal, partner David Handler and associate Tony Ray Meyer-Mangione discuss the rationale for using a specific property approach for sequestered, traceable assets and the fractional approach for commingled assets.
In a grantor trust, the grantor (e.g., the creator of the trust or a person who made a gratuitous transfer to the trust) must report on her personal income tax return all items of income, loss, deduction and credit attributable to the portion of the trust which she is deemed to own under the grantor trust rules of Sections 671 – 679 of the Internal Revenue Code (the Code). This is straightforward when the trust is treated as wholly-owned by a single grantor. But when multiple grantors are treated as owners of different portions of the trust (a “multiple-grantor trust”), how should those grantors divide the income of the trust among themselves? Treasury Reg. Section 1.671-3 provides three possibilities for how ownership could be determined: by a fractional share (the “fractional approach”), by a pecuniary amount, or by tracing specific property (the “specific property approach”). However, Reg. Section 1.671-3 is generally drafted from the perspective of a single grantor who is treated as owner with respect to a portion of the trust. The regulations do not provide clear guidance as to how to determine the ownership of a multiple-grantor trust, and it is not certain the circumstances in which the IRS or a court would view each of these approaches as the more (or only) appropriate method for determining how income should be allocated among grantors.
Given the tax policy justifications for the grantor trust rules and based on analogous case precedent, we discuss the rationale for using the specific property approach for sequestered, traceable assets and the fractional approach for commingled assets.
Scenarios 1 and 2
Scenario 1. First, consider an irrevocable trust to which two individuals (John and Jane) transfer assets: stock from John (valued at 50% of the trust assets) and LLC units from Jane (also valued at 50%). Due to a substitution power held by John and Jane over all the trust property, both are treated as grantors under Code Section 675. How should the trust’s income be allocated among them?
In this scenario, the regulations provide two options: trace income to the specific property each grantor contributed, or determine a fractional share of the trust each grantor is deemed to own. If using the specific property approach, John will include on his tax return income attributable to the stock, while Jane will include income attributable to the LLC units. If using the fractional approach, each instead reports 50% of the trust’s overall income, evenly dividing the stock income and LLC income. In the fractional approach, when the respective asset values change, the fraction of a grantor’s ownership (e.g., 50%) does not – however, in the specific property approach, a grantor’s share of the trust’s taxes fluctuates with the income and gains from the asset he or she contributed.
Scenario 2. Consider Example 7 of Reg. Section 1.671-2(e)(6). Here, person A is the sole grantor of a trust. When the trust’s stock is worth $100,000, Person C sells property worth $1,000,000 to the trust in a part-sale ($100,000), part-gratuitous transfer ($900,000). Person C is treated as the grantor with respect to the $900,000 portion of the trust (the gratuitous transfer portion), and Person A is treated as the grantor with respect to the $100,000 portion.
Assume, as in Scenario 1, that due to a substitution power over all the trust property, both are treated as grantors under Code Section 675. However, Example 7 does not state whether Person A should be treated as the grantor of (1) the specific property she contributed or (2) a 10% undivided fractional interest.
Historical Policy and the ‘Estate of Bell’
The legislative history of the grantor trust rules ties them to the Helvering v. Clifford decision and the necessity to protect the progressive income tax structure. When the grantor trust rules were codified in 1954, trusts and individuals had similar tax rates and identical maximum rates, so taxpayers could divide income among many trusts to reduce a taxpayer’s effective tax rate (i.e., income-shifting). The Senate Committee on Finance stated in their 1924 explanatory report that “trusts have been used to evade taxes by means of provisions allowing the distribution of the income to the grantor or its use for his benefit. The purpose of [Section 219, precursor to Section 676] is to stop this evasion.” Thus, the income-shifting by a grantor to a trust with lower tax brackets was resolved by treating grantors as owners of the property they contributed to such trusts.
In the 1976 case Estate of Bell, the Tax Court addressed this allocation issue in the context of Code Section 2038. In Bell, the IRS argued that the value of the portion of the trust includible in the decedent’s gross estate should be determined by a fractional share. The Tax Court noted that “where property so transferred by the several grantors of a trust is commingled with other trust property and cannot be identified, the formula advocated by [the IRS] may have merit.” The Tax Court further held that “if specific property transferred by a decedent is capable of identification, however, the value of such specific property should be included in the gross estate [and the value of specific assets contributed by others] should be excluded in making the required allocations.”
(Note that Bell involved estate tax under a rule that applies to contributed property in which the donor retained an interest, not a grantor trust. However, the logic of Bell offers a useful analogue to the treatment of multiple-grantor trusts.)
Thus, in extending Bell, we see two different approaches merited by our scenarios. In Scenario 1, it is possible to trace the specific property that John and Jane each transferred to the trust. In Bell, the Tax Court reasoned that since the decedent’s specific property appreciated less than other trust assets, attributing to her a fractional share of the property resulted in unfair inclusion in her estate. The same logic applies to John and Jane. When the fractional approach is used, if John’s stock earns significantly more income than Jane’s LLC units, Jane faces significant income liability on account of property she did not transfer to the trust. The specific property approach is more equitable here under the logic of Bell. Moreover, the specific property approach aligns with the historical policy rationale that grantors should not engage in income-shifting (e.g., between grantors on account of contributions to a multiple-grantor trust rather than separate, wholly-owned trusts).
In Scenario 2, following the sale and gratuitous transfer by Person C, the trust has $1 million of total assets. Thus, since there is not a difference in the type of assets deemed owned by Person A and Person C (a “commingling” under the terms of Bell), the fractional approach may have more merit in Scenario 2.
Is There a Choice Between Approaches?
Given the policy goal of avoiding income-shifting, can taxpayers choose the fractional approach when the tracing of specific property is feasible? Under existing guidance, the answer is not clear—and there is potential for abuse. If the fractional approach is used when specific, traceable assets are contributed to a trust (as in Scenario 1), it would allow grantors who are taxable at different marginal tax brackets and who live in different states to contribute assets with different tax bases to a single multiple-grantor trust, thereby shifting taxation among them solely based on the value of the assets contributed. For example, say John lives in California and contributes $1 million of stock with no basis, while Jane lives in Texas and contributes $1 million of LLC units with full basis. If the fractional approach is applied, then when the stock is sold, John and Jane will each be taxable on only half the capital gain. If John sold the stock himself, he would pay tax on the entire gain. Thus, the amount and bearer of the federal and state taxes is reduced and shifted through the trust. Such a result would not occur in other contexts (such as a partnership), and likely was not a contemplated result of the fractional approach. In fact, it would thwart the original purpose of the grantor trust rules: the prevention of income-shifting among taxpayers subject to different tax rates.
Conclusion
The specific property approach requires tracing assets to ensure items of income and deduction specific to such property are allocated to the grantors who contributed them. Moreover, the income, proceeds and reinvestment of such property needs to be traced to continue fairly allocating income among the grantors. If such property later becomes commingled to the extent such identification is not feasible (e.g., upon the sale of such property for cash which is not segregated), then the fractional approach likely becomes the best approach. However, IRS guidance and court precedent is largely silent on this issue, requiring a careful facts and circumstances analysis to ensure the most appropriate method is applied in allocating income among grantors of multiple-grantor trusts.
